Hacker News new | ask | show | jobs
by xoa 900 days ago
>Which... I largely don't care about. I understand why it's super important for academics, but in my book it's not a con or scam. It's accurate information.

This is not correct on a few levels, at least in the context of science. At the most basic, you're engaging in circular reasoning. You're accepting it's "accurate information" as a truism when the point of science is to discover what is actually "accurate information" in the absence of any oracles. Someone who is plagiarizing doesn't actually know whether what they're plagiarizing is accurate or not, by definition they haven't done the work. They don't know how it all connects together, and not only might what they're copying be wrong, they're more likely to introduce errors of context and omit qualifiers.

Tying into that is the issue of meta-information as well. One of the core foundations of science and assessing whether information is accurate or not is replication. A second/third/fourth/etc researcher independently reaching even 100% identical results is itself new information each time, even if conclusion is the same. More independent replication raises the chance of signal and decreases the chance that it was noise, some unaccounted for variables unique to a given lab or researcher. Everyone makes mistakes, but even with zero mistakes low probability things can happen in any single given experiment/study/place. Diverse distributed replication is a basic way to help discover/dismiss that.

A plagiarist in research is therefore, at a bare minimum, always engaged in a con/scam: they're claiming they have independently produced a result, which then adds to the weight and other people will be more likely to depend on. They have not.

Of course, they've also conned/scammed whatever money/time/resources anyone else contributed to them with the expectation of independent work and thus at a minimum new replication information. They took that, and then didn't follow through. It's fraud. And there is opportunity cost there since those are a zero-sum game, the resources that went to funding a plagiarist could have instead gone to fund someone honest who could produce something with actual ROI as expected.

2 comments

Every single case of plagiarism can be fixed with some combination of quotation marks, proper citations, and rephrasing of relevant text. It is never a claim to have independently reproduced a result.

If a scientist claims to have independently reproduced an experimental result - and they haven't - that is outright fraud. It doesn't matter if they describe that experiment in original words, with proper citations and quotation marks/blocks.

You're making a lot of assumptions about the nature of the plagiarized content.

If plagiarizing is bad because the copied info might be wrong, then original research is bad because it might be wrong.

>If plagiarizing is bad because the copied info might be wrong, then original research is bad because it might be wrong.

No, that's not what I wrote at all. Original honest research might be wrong and that's completely fine and understood, that's the entire point of replication! But if someone lies and says they've independently replicated it when they just plagiarized that's fraud and wrong. Telling the world original research has been replicated when it hasn't been is different then a tentative first result. Depending on the area nobody else may then attempt replication for awhile, assuming it's already been done and instead waste resources trying to build upon it only to later discover it was wrong. The original team has also been robbed of independent review they may have otherwise gotten, and themselves move on only to much later have the rug pulled out from under them. Depending on the nature of what caused the original research to be incorrect it may have further ripple effects. Say some piece of equipment or new instrument has some subtle flaw that a true replication attempt would have discovered, but instead it then keeps getting used and causing other experiments to go off.

And this is all in the generous case where the plagiarist does at least pretend to have the same result as what they're copying, vs hacking it into something else and generating total garbage completely.